Search bioRxiv⌕ Search

bioRxiv · 10.1101/2025.07.31.667850

Integrative immune subtyping of HNSCC reveals clinically relevant phenotypes and treatment-associated transitions

Abstract

Head and neck squamous cell carcinoma (HNSCC) exhibits profound heterogeneity in clinical presentation, treatment response, and immune landscape. While prior classification systems have identified molecular and immune subtypes in this disease, their applicability to real-world clinical settings remains restricted to small, homogeneous cohorts and limited by lack of multimodal data integration and interpretation. We performed integrated multi-omics analysis including transcriptomic, genomic (copy number, single-nucleotide variants) on 1,149 tumors from 1,102 HNSCC patients across treatment settings. Using the Similarity Network Fusion (SNF) algorithm, we defined immune subtype clusters (ISCs) based on the full immune gene landscape. These clusters were characterized using mutational, transcriptional, and immune cell enrichment analyses, and mapped to hypoxia and traditional subtypes. Associations with clinical outcomes, including progression-free survival, were evaluated across first-line and post-metastatic treatment settings. Four distinct immune subtype clusters (ISC1-ISC4) were identified: ISC1: immune-cold and EMT-enriched; ISC2: immune activated; ISC3: mixed immune-regulatory and stromal-enriched phenotype; and ISC4: immunosuppressed. Distinct treatment response patterns were observed across subtypes in subjects treated with checkpoint inhibitors, chemotherapy, and combination regimens. 44 Patients with matched pre/post treatment tumors revealed treatment-associated transitions between immune subtypes: checkpoint inhibitor treatment enriched for immune activation, while chemotherapy treatment enriched for immunosuppressive signaling pathways. This study provides a clinically relevant immune subtyping framework for HNSCC based on real-world, multi-omics data. These subtypes reflect dynamic tumor-immune states and associated with treatment response and survival, supporting their use in guiding immune-based therapy in HNSCC.

Source connections

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Khatri, I., de Souza, T., van Asten, S., Belete, M., Braga, F. V., Jongmans, M., Sridhar, S., Higgs, B. W., Kolder, I.. 2025-08-02. Integrative immune subtyping of HNSCC reveals clinically relevant phenotypes and treatment-associated transitions. https://doi.org/10.1101/2025.07.31.667850

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

Common viral infections seed regionally distinct resident memory T cells in the human CNS

T cells persist in the central nervous system (CNS) and can drive both protection and neurological disease. How these cells are organized in humans and what they recognize is largely unknown. Here, we profiled CD8 T cells across anatomically distinct CNS regions, obtained through on-site autopsies and temporal lobe resection surgeries, using single-cell RNA sequencing, paired T cell receptor sequencing, and DNA-barcoded tetramers. Resident memory T cells (TRM) specific for Epstein-Barr virus, cytomegalovirus, influenza A, and SARS-CoV-2 were identified across CNS compartments. Anatomical location was the strongest correlate of TRM cell state, with leptomeningeal cells adopting a cytokine-poised TRM program, whereas brain TRM cells were transcriptionally restrained. Cells of the same clonotype spanned tissues yet adopted local transcriptional states. Viral specificity added another layer of TRM heterogeneity with GZMK/GZMA-expressing EBV-specific populations and interferon-stimulated gene signatures in SARS-CoV-2 and Influenza A-specific cells. The human CNS thus harbors regionally distinct CD8+ TRM shaped by common viral exposures.

immunology↗

A regulatory T cell signature provides a shared molecular basis for the therapeutic window of opportunity in rheumatic disease

Rheumatic diseases, including rheumatoid arthritis (RA), spondyloarthritis (SpA) and osteoarthritis (OA), show distinct phenotypes yet respond to overlapping therapies, implicating shared immune mechanisms. In the Transimmunom cohort, we profiled peripheral blood from 240 individuals (47 healthy, 44 OA, 91 RA, 58 SpA) across deep immunophenotyping, immunoproteomics and Treg-Teff transcriptomics. Single-layer analyses revealed broader Treg than Teff remodeling, along with a shared pattern of reduced activated Tregs and expanded Helios+ Tregs across all diseases, alongside a decrease in functional Treg subpopulations, including CTLA4+ and CD45RA- Tregs. In RA specifically, LAG3+ Tregs were also expanded. Combining omics layers outperformed single-layer approaches for disease classification. Among individual layers, Treg transcriptomes were most discriminative, and integration uncovered disease-specific programs. Unsupervised clustering identified a cross-disease cluster independent of activity, treatment and age, mapping to early disease (<= years) and dominated by a Treg dysfunction-associated program. These results provide a biological rationale for the therapeutic "window of opportunity" concept and duration-stratified Treg-directed trials.

immunology↗

Inhibitory Fc Receptor sets a time limit on macrophage response to IgG

Antibodies engage both activating Fc Receptors and the inhibitory receptor Fc{gamma}RIIB. Why macrophages need a dedicated inhibitory receptor rather than simply tuning activating receptor signaling is unclear. Using DNA-based chimeric receptors and in silico modeling, we independently controlled activating and inhibitory Fc Receptors. We found that Fc{gamma}RIIB imposed a time limit on macrophage phagocytosis and ERK signaling. The time limit is due to activating Fc Receptors converting PI(4,5)P2 to PI(3,4,5)P3, which is subsequently converted to PI(3,4)P2 by Fc{gamma}RIIB. This leads to a pulse of active signaling, which is sufficient for phagocytosis of small bacteria-sized targets but not phagocytosis of large targets and TNF secretion. Unlike engaging Fc{gamma}RIIB, reducing activating Fc Receptor signaling decreased initiation of phagocytosis, the speed of PI(3,4,5)P3 generation, and the amplitude of ERK signaling. Our results demonstrate that Fc{gamma}RIIB controls the duration of IgG signaling, while the activating Fc Receptors control sensitivity.

immunology↗